In most instances, economies do not operate at their productive and technical upper bound. Developing countries do not have to invent new techniques. They can borrow techniques and types of capital that already exist. (Of course, in detail, some research and experimentation is frequently necessary to adapt broadly known techniques to specific local conditions.) In view of these considerations, the finding that traditional inputs account for only a small proportion of the growth that takes place is hardly surprising. The old Malthusian argument that additions to the population come into the world with additional hands but without the additional capital or land necessary to produce at the same level as their forebears is not entirely true. The nurture and educational system can create to some degree the additional capital necessary. Whether this "human capital" is adequate or not depends upon the rate of transmission of known and new skills, and the simultaneous introduction of other types of capital into the population. (The word skill is used in its broadest possible sense in this context.) The rate of growth of physical capital may be to some extent a function of the growth rate of human capital. The basic argument is neither pro- nor anti-Malthusian. Rather it suggests that the traditional approach misses to a considerable degree the fundamental processes which determine whether or not given rates of population growth are adverse to economic growth.
The point emphasized in this paper is that the research of the last 15 years shows that whereas the output that results from traditional inputs may not be entirely trivial, it is nevertheless not nearly as important as the contribution to output that results from nontraditional inputs. Many of the nontraditional inputs have an elusive quality about them. They cannot be handled from an analytical viewpoint as easily as the traditional ones. The basic conjecture of this paper is that the assumption of a one-to-one correspondence between inputs and outputs is no longer tenable once one gives primary importance to nontraditional inputs. What is new is the rather persuasive evidence that the nontraditional inputs are usually more significant than the traditional ones, and hence the relations between population growth and nontraditional inputs should in most cases become central to the analysis of the "population-resources" problem.
For the most part, this paper is only suggestive. The overall results are inconclusive since we really do not know very much about the economic magnitudes of the demographic factors that are detrimental to economic growth, as against those that are helpful. This lack is especially true of the effects other than the replacement effect. The replacement-effect type of argument is of interest since it suggests that even in developing countries, there may be situations and periods for which relatively high rates of population growth may involve some demographic effects that are helpful to economic growth. Whether the beneficial effects are ever the predominant ones is hard to say, but it is a possibility that cannot be entirely ignored. In examin-